- 12 Jan, 2015 2 commits
-
-
David Bigagli authored
-
David Bigagli authored
-
- 10 Jan, 2015 1 commit
-
-
Morris Jette authored
Remove GRES spec from burst_buffer.conf There is no logic in the generic plugin to support this and the cray plugin does not get the information from the config file. Change allow/deny user separator from colon to comma Confirm AllowUsers for swap and gres space Make sure nodes not shared on cray system Minor changes in several other places
-
- 09 Jan, 2015 8 commits
-
-
Morris Jette authored
-
Danny Auble authored
Conflicts: META NEWS src/plugins/mpi/pmi2/tree.c
-
Danny Auble authored
-
Danny Auble authored
-
Morris Jette authored
-
Danny Auble authored
-
Danny Auble authored
-
Danny Auble authored
This is needed for setups like this TaskPlugin = affinity TaskPlugin = task/affinity,task/cgroup TaskPlugin = affinity,cgroup
-
- 08 Jan, 2015 10 commits
-
-
Daniel Pou authored
Modify contribs/sgi Makefile and netloc_to_topology.c to use configure parameters related to netloc install
-
Michael Raymond authored
Document how to build an SGI hypercube configuration file.
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
The "full info" display was not working Duplicate records sometimes were displayed The job ID information might have been trucated Add comment that burst_buffer memory leak check is needed
-
Morris Jette authored
-
Morris Jette authored
Fix job array bug
-
Brian Christiansen authored
-
Morris Jette authored
This gets BB working with job arrays
-
Brian Christiansen authored
Bug 1352
-
- 07 Jan, 2015 9 commits
-
-
Morris Jette authored
Add support for job arrays (mostly working)
-
Brian Christiansen authored
Bug 1352
-
Danny Auble authored
-
Danny Auble authored
-
David Bigagli authored
Slurm 14.03 nccs
-
Aaron Knister authored
-
Rémi Palancher authored
Intel MPI, on MPI jobs initialisation through PMI, uses to call PMI_KVS_Put() many many times from task at rank 0, and each on these call is followed by PMI_KVS_Commit(). Slurm implementation of PMI_KVS_Commit() imposes a delay to avoid DDOS on original srun. This delay is proportional to the total number. It could be up to 3 secs for large jobs for ex. with 7168 tasks. Therefore, when Intel MPI calls PMI_KVS_Commit() 475 times (mesured on a test case) from task at rank 0, 28 minutes are spent in delay function. All other tasks in the job are waiting for a PMI_Barrier. Therefore, there is no risk for a DDOS from this single task 0. The patch alters the delaying time calculation to make sure task at rank 0 will does not be delayed. All other tasks are globally spreaded in the same time range as before.
-
Aaron Knister authored
-
Artem Polyakov authored
-
- 06 Jan, 2015 10 commits
-
-
David Bigagli authored
-
Morris Jette authored
Move a couple more variable definitions to the top of a block
-
Morris Jette authored
Added Makefile for contribs/sgi file. Moved hypercube symbol definitions from select/linear to common. Minor format changes for consistency with other Slurm code. Moved a variable definition (l_distance) to start of code block to avoid error with some compilers. Fix for possible uninitialized variable use (leftover_nodes).
-
Michael Raymond authored
-
Christopher Samuel authored
Misplaced ";" resulted in bad tracking of BB user space usage
-
Morris Jette authored
-
Morris Jette authored
Ammendment to commit 744f114b
-
Morris Jette authored
-
David Bigagli authored
-
Morris Jette authored
-